DeployWhere

RunPod

Cloud GPU infrastructure for AI development, training, inference and multi-node workloads.

No reviews yet. Write the first.
Visit RunPod

RunPod is a GPU and AI hosting provider based in United States. It operates European infrastructure in Netherlands, Sweden and Romania and other locations. There is no free tier. As a US company it falls under the US CLOUD Act. Pricing on this page is published only after a human has reviewed it.

PricingPros and consSpecificationsAboutReviews
Company based in
United States
European regions
NetherlandsSwedenRomaniaCroatia
US CLOUD Act
US company
GDPR agreement
Available
Free tier
None
VAT
Shown excluding VAT
RunPod website
Website image published by RunPod.

Pricing

42 plans
  • Pods: B300
    GPU rental is billed by the second or millisecond.
    Monthly
    about 5759.7 USD$7.89/hr
    vCPU
    32
    Memory
    251 GB
    Storage
    Not listed
  • Pods: H200
    GPU rental is billed by the second or millisecond.
    Monthly
    about 3350.7 USD$4.59/hr
    vCPU
    24
    Memory
    276 GB
    Storage
    Not listed
  • Pods: B200
    GPU rental is billed by the second or millisecond.
    Monthly
    about 4956.7 USD$6.79/hr
    vCPU
    28
    Memory
    283 GB
    Storage
    Not listed
  • Pods: RTX Pro 6000
    GPU rental is billed by the second or millisecond.
    Monthly
    about 1525.7 USD$2.09/hr
    vCPU
    16
    Memory
    188 GB
    Storage
    Not listed
  • Pods: H100 NVL
    GPU rental is billed by the second or millisecond.
    Monthly
    about 2328.7 USD$3.19/hr
    vCPU
    16
    Memory
    94 GB
    Storage
    Not listed
  • Pods: H100 PCIe
    GPU rental is billed by the second or millisecond.
    Monthly
    about 2109.7 USD$2.89/hr
    vCPU
    16
    Memory
    188 GB
    Storage
    Not listed
  • Pods: H100 SXM
    GPU rental is billed by the second or millisecond.
    Monthly
    about 2401.7 USD$3.29/hr
    vCPU
    20
    Memory
    125 GB
    Storage
    Not listed
  • Pods: A100 PCIe
    GPU rental is billed by the second or millisecond.
    Monthly
    about 1014.7 USD$1.39/hr
    vCPU
    8
    Memory
    117 GB
    Storage
    Not listed
  • Pods: A100 SXM
    GPU rental is billed by the second or millisecond.
    Monthly
    about 1160.7 USD$1.59/hr
    vCPU
    16
    Memory
    125 GB
    Storage
    Not listed
  • Pods: L40S
    GPU rental is billed by the second or millisecond.
    Monthly
    about 722.7 USD$0.99/hr
    vCPU
    16
    Memory
    94 GB
    Storage
    Not listed
  • Pods: RTX 6000 Ada
    GPU rental is billed by the second or millisecond.
    Monthly
    about 613.2 USD$0.84/hr
    vCPU
    10
    Memory
    167 GB
    Storage
    Not listed
  • Pods: A40
    GPU rental is billed by the second or millisecond.
    Monthly
    about 321.2 USD$0.44/hr
    vCPU
    9
    Memory
    50 GB
    Storage
    Not listed
  • Pods: L40
    GPU rental is billed by the second or millisecond.
    Monthly
    about 598.6 USD$0.82/hr
    vCPU
    8
    Memory
    94 GB
    Storage
    Not listed
  • Pods: RTX A6000
    GPU rental is billed by the second or millisecond.
    Monthly
    about 386.9 USD$0.53/hr
    vCPU
    9
    Memory
    50 GB
    Storage
    Not listed
  • Pods: RTX 5090
    GPU rental is billed by the second or millisecond.
    Monthly
    about 722.7 USD$0.99/hr
    vCPU
    9
    Memory
    35 GB
    Storage
    Not listed
  • Pods: L4
    GPU rental is billed by the second or millisecond.
    Monthly
    about 357.7 USD$0.49/hr
    vCPU
    12
    Memory
    50 GB
    Storage
    Not listed
  • Pods: RTX 3090
    GPU rental is billed by the second or millisecond.
    Monthly
    about 365 USD$0.50/hr
    vCPU
    16
    Memory
    125 GB
    Storage
    Not listed
  • Pods: RTX 4090
    GPU rental is billed by the second or millisecond.
    Monthly
    about 540.2 USD$0.74/hr
    vCPU
    6
    Memory
    41 GB
    Storage
    Not listed
  • Pods: RTX A5000
    GPU rental is billed by the second or millisecond.
    Monthly
    about 197.1 USD$0.27/hr
    vCPU
    9
    Memory
    25 GB
    Storage
    Not listed
  • Serverless: B300
    Inference worker usage pricing.
    Monthly
    about 7285.4 USD$9.98/hr
    vCPU
    Not listed
    Memory
    Not listed
    Storage
    Not listed
  • Serverless: B200
    Inference worker usage pricing.
    Monthly
    about 6307.2 USD$8.64/hr
    vCPU
    Not listed
    Memory
    Not listed
    Storage
    Not listed
  • Serverless: H200
    Inference worker usage pricing.
    Monthly
    about 4328.9 USD$5.93/hr
    vCPU
    Not listed
    Memory
    Not listed
    Storage
    Not listed
  • Serverless: RTX 6000 Pro
    Inference worker usage pricing.
    Monthly
    about 2547.7 USD$3.49/hr
    vCPU
    Not listed
    Memory
    Not listed
    Storage
    Not listed
  • Serverless: H100
    Inference worker usage pricing.
    Monthly
    about 3496.7 USD$4.79/hr
    vCPU
    Not listed
    Memory
    Not listed
    Storage
    Not listed
  • Serverless: A100
    Inference worker usage pricing.
    Monthly
    about 1985.6 USD$2.72/hr
    vCPU
    Not listed
    Memory
    Not listed
    Storage
    Not listed
  • Serverless: L40, L40S, 6000 Ada, MIG 48GB
    Inference worker usage pricing.
    Monthly
    about 1277.5 USD$1.75/hr
    vCPU
    Not listed
    Memory
    Not listed
    Storage
    Not listed
  • Serverless: A6000, A40
    Inference worker usage pricing.
    Monthly
    about 890.6 USD$1.22/hr
    vCPU
    Not listed
    Memory
    Not listed
    Storage
    Not listed
  • Serverless: 5090
    Inference worker usage pricing.
    Monthly
    about 1153.4 USD$1.58/hr
    vCPU
    Not listed
    Memory
    Not listed
    Storage
    Not listed
  • Serverless: RTX PRO 4500 Blackwell
    Inference worker usage pricing.
    Monthly
    about 839.5 USD$1.15/hr
    vCPU
    Not listed
    Memory
    Not listed
    Storage
    Not listed
  • Serverless: 4090
    Inference worker usage pricing.
    Monthly
    about 803 USD$1.10/hr
    vCPU
    Not listed
    Memory
    Not listed
    Storage
    Not listed
  • Serverless: L4, A5000, 3090, MIG 24GB
    Inference worker usage pricing.
    Monthly
    about 503.7 USD$0.69/hr
    vCPU
    Not listed
    Memory
    Not listed
    Storage
    Not listed
  • Serverless: A4000, A4500, RTX 4000, RTX 2000
    Inference worker usage pricing.
    Monthly
    about 423.4 USD$0.58/hr
    vCPU
    Not listed
    Memory
    Not listed
    Storage
    Not listed
  • Clusters: H200 SXM
    Multi-GPU cluster pricing with no commitments.
    Monthly
    about 3146.3 USD$4.31/hr
    vCPU
    Not listed
    Memory
    Not listed
    Storage
    Not listed
  • Clusters: A100 SXM
    Multi-GPU cluster pricing with no commitments.
    Monthly
    about 1306.7 USD$1.79/hr
    vCPU
    Not listed
    Memory
    Not listed
    Storage
    Not listed
  • Clusters: L40S
    Monthly
    Contact sales
    vCPU
    Not listed
    Memory
    Not listed
    Storage
    Not listed
  • Clusters: H100 SXM
    Monthly
    Contact sales
    vCPU
    Not listed
    Memory
    Not listed
    Storage
    Not listed
  • Clusters: B200
    Monthly
    Contact sales
    vCPU
    Not listed
    Memory
    Not listed
    Storage
    Not listed
  • Reserved Clusters: H200 SXM
    Dedicated clusters with guaranteed availability and SLA-backed uptime. Terms shown for 1mo, 3mo, 6mo, 12mo and 12mo+.
    Monthly
    Contact sales
    vCPU
    Not listed
    Memory
    Not listed
    Storage
    Not listed
  • Reserved Clusters: A100 SXM
    Dedicated clusters with guaranteed availability and SLA-backed uptime. Terms shown for 1mo, 3mo, 6mo, 12mo and 12mo+.
    Monthly
    Contact sales
    vCPU
    Not listed
    Memory
    Not listed
    Storage
    Not listed
  • Reserved Clusters: L40S
    Dedicated clusters with guaranteed availability and SLA-backed uptime. Terms shown for 1mo, 3mo, 6mo, 12mo and 12mo+.
    Monthly
    Contact sales
    vCPU
    Not listed
    Memory
    Not listed
    Storage
    Not listed
  • Reserved Clusters: H100 SXM
    Dedicated clusters with guaranteed availability and SLA-backed uptime. Terms shown for 1mo, 3mo, 6mo, 12mo and 12mo+.
    Monthly
    Contact sales
    vCPU
    Not listed
    Memory
    Not listed
    Storage
    Not listed
  • Reserved Clusters: B200
    Dedicated clusters with guaranteed availability and SLA-backed uptime. Terms shown for 1mo, 3mo, 6mo, 12mo and 12mo+.
    Monthly
    Contact sales
    vCPU
    Not listed
    Memory
    Not listed
    Storage
    Not listed

Pros and cons

What works

  • Per-second or millisecond compute billing is stated.
  • No contracts, minimums or egress fees are stated for on-demand GPU rentals.
  • Custom Docker containers, API access, CLI and SDKs are supported.
  • Pods, Serverless and Clusters cover persistent, autoscaling and multi-node workloads.

What to watch

  • Specific datacenter countries are not listed.
  • GPU, Serverless and cluster prices are primarily hourly usage rates rather than monthly plans.
  • Some cluster and all reserved cluster prices require contacting sales.
  • Storage is charged separately, including different running and idle rates for some storage types.
  • Spot instances can be evicted when demand spikes.

Specifications

Regions
US, Europe, Asia, Australia
Supported runtimes
Linux, PyTorch, TensorFlow, JAX, ONNX, CUDA toolkits, NCCL, Horovod, vLLM, TGI, custom Docker containers
Egress policy
No egress fees are stated for GPU instances and storage.
Data residency guarantee
false
Certifications
SOC 2 Type 2, SOC 3, HIPAA, ISO 27001
Uptime sla
99.99% uptime guarantee
Support
Enterprise support, sales and compliance reviews are available through direct contact; hours and paid support tiers are not stated.

About RunPod

Runpod provides dedicated GPU Pods, autoscaling Serverless inference and multi-node Clusters. Pods expose the container, storage, GPU and runtime environment, while compute is priced with hourly figures and billed by the second or millisecond. The pages list 31 global regions but do not name specific countries, and storage is charged separately with no ingress or egress fees stated.

Reviews

None yet

No approved reviews yet. Every review is read by a person before it appears, and only approved reviews count toward a rating.

RunPod head to head

RunPod against Vast.ai
Price, regions and free tier side by side

Other gpu and ai hosting

See all
DataCrunch
Finland
Exoscale GPU
Switzerland
IONOS AI
Germany
Nebius
Netherlands
OVHcloud AI
France
Scaleway GPU Instances
France

Something here out of date?

Every figure on this page was checked by a person, and providers change their plans without telling anybody. Say what moved and it gets checked again.

Work at RunPod? Claim this listing to correct the description and add screenshots. Verification is a confirmation link to an address on your own domain. Pricing still goes through the same check as every other listing here.